Content uploaded by Anine Nordstrøm
Author content
All content in this area was uploaded by Anine Nordstrøm on Nov 07, 2022
Content may be subject to copyright.
Original Research
Association of Training and Game Loads
to Injury Risk in Junior Male Elite
Ice Hockey Players
A Prospective Cohort Study
Anine Nordstrøm,*
†‡
MD, Roald Bahr,
†
MD, PhD, Lena K. Bache-Mathiesen,
†
,
Ben Clarsen,
†§
PT, PhD, and Ove Talsnes,
‡k
MD, PhD
Investigation performed at Oslo Sports Trauma Research Center, Department of Sports Medicine,
Norwegian School of Sports Sciences, Oslo, Norway
Background: Training and game loads are potential risk factors of injury in junior elite ice hockey, but the association of training
and game loads to injuries is unknown.
Purpose: To investigate the association of chronic training and game loads to injury risk in junior male elite ice hockey players.
Study Design: Cohort study; Level of evidence, 2.
Methods: In this prospective cohort study, we monitored all health problems among 159 male junior ice hockey players (mean age,
16 years; range, 15-19 years) at sports-specific high schools during the 2018-2019 school year. Players reported their health
problems every week using the Oslo Sports Trauma Research Center Overuse Questionnaire on Health Problems (OSTRC-H2).
The number of training sessions and games was reported for 33 weeks. We calculated the previous 2-week difference in training/
game loads as well as the cumulative training/game loads of the previous 2, 3, 4, and 6 weeks and explored potential associations
between training/game loads and injury risk using mixed-effects logistic regression.
Results: The players reported 133 acute injuries, 75 overuse injuries, and 162 illnesses in total, and an average of 8.8 (SD ±3.9)
training sessions and 0.9 (SD ±1.1) games per week. We found no association between the difference of the two previous weeks or
the previous 2- 3- and 4-week cumulative, training or game load and acute injuries, nor the difference of the two previous weeks, or
the previous 4- and 6-week cumulative, training or game load and overuse injuries (OR, *1.0; P>.05 in all models).
Conclusion: In the current study of junior elite ice hockey players, there was no evidence of an association between cumulative
exposure to training/game loads and injury risk.
Keywords: load; overuse injuries; ice hockey; epidemiology; injury prevention; junior injuries; adolescent injuries
Ice hockey is a contact sport in which players frequently
collide, with each other, boards, or goals, and get hit by
sticks and pucks; injuries are prevalent. Injury risk is sub-
stantially higher during games (14.73 per 1000 athlete-
exposures
45
; 30.3-49.7
57
and 96.1
58
per 1000 player game
hours) than during training sessions (2.52 per 1000 athlete-
exposures
45
; 3.9 per 1000 player practice hours
58
). The risk
of injury increases with age.
57,60
Nordstrøm et al
52
found an
average weekly injury prevalence of 20%among junior elite
players, split equally between acute and overuse injuries.
Injuries have a significant impact on player and team
performance.
37
Injury cause is multifactorial and involves both extrinsic
and intrinsic risk factors,
47
and recent research has sug-
gested that poor workload management may contribute to
injuries and illnesses in sports.
26,28
Across different sports,
training and game load have been suggested as risk factors
for health problems,
1
and several studies have reported an
association between training load and injury.
15,24,29,41,44,48
Studies in rugby, basketball, and Australian football have
shown that the greatest incidence of illness and injury is
observed when current training loads are greatest.
2,19,20,33
Several studies have found that a large percentage of
injuries are associated with rapid changes or spikes in
weekly loads
31,34,55
and that competition congestion leads
to increased risk of injury.
10,27
Previous studies have also
found a relationship between relative change in training
load, measured by the acute/chronic workload ratio
The Orthopaedic Journal of Sports Medicine, 10(10), 23259671221129646
DOI: 10.1177/23259671221129646
ªThe Author(s) 2022
1
This open-access article is published and distributed under the Creative Commons Attribution - NonCommercial - No Derivatives License (https://creativecommons.org/
licenses/by-nc-nd/4.0/), which permits the noncommercial use, distribution, and reproduction of the article in any medium, provided the original author and source are
credited. You may not alter, transform, or build upon this article without the permission of the Author(s). For article reuse guidelines, please visit SAGE’s website at
http://www.sagepub.com/journals-permissions.
(ACWR), and injury risk.
34,13,35,36
However, this concept’s
methodology has been criticized recently.
38-40,43
In a small study regarding collegiate ice hockey, Musta-
pich and Koehle
50
found that players with greater training
load experienced greater odds of injury compared with
players with lower training loads. The association between
training/game load and injury in junior elite ice hockey has
not been studied.
The purpose of our study was to investigate the associa-
tion between chronic training/game load and acute and
overuse injuries in junior elite ice hockey players.
METHODS
Study Design and Participants
This was a prospective cohort study on junior elite ice
hockey players in 5 sports-specific high schools and 2 clubs
during the 2018-2019 school year. The data collection meth-
ods used are the same as those recently applied in 2 studies
in Norwegian ice hockey.
52,53
The prevalence and burden of
all health problems are reported in a separate paper.
52
The protocol for this study received ethics committee
approval, and all players provided written informed con-
sent to participate in the study. For those younger than
18 years, written consent was signed by their parents. All
participants approved access to their data for their school
staff and physician.
Recruitment and Inclusion Criteria
We included 6 private specialized sports academy high
schools that offer elite sports programs to students who
want to combine sports on a high level with a college-
entry academic program. The principal investigator (A.N.)
contacted the schools, their management, and their coaches
by email and telephone in January 2018 and held meetings
with all schools providing information about the study dur-
ing the winter-spring of 2018. The schools had limited med-
ical support: one of the schools had a physician, while the
rest had dedicated school staff who contacted medical per-
sonnel when needed. We informed the players about the
study during a meeting at their school at the start of the
school year. One school preferred that the study be con-
ducted through their local affiliated club. We enrolled 2
teams (under 18 years and under 21 years) in the study.
In the 5 schools and 1 club included, 3 players declined the
invitation to participate in the study, 2 players did not
respond in the system after they agreed to participate, and
4 players stopped reporting after 18 weeks. A total of 23
players dropped out throughout the year because of change
of school, change of team, or unknown reasons. A female
player was excluded from the analyses. A total of 47 players
reported trainings and games <50%of all possible weeks
and were excluded from the study. The final study sample
consisted of 159 players.
Injury and Illness Data Collection
The injury and illness registration and collection procedures
have been reported previously.
52,53
Surveillance was con-
ducted using an online platform designed to collect injury
and illness data from athletes using the Oslo Sports
Trauma Research Center Overuse Questionnaire on
Health Problems (OSTRC-H2).
16-18
The OSTRC-H2 was
distributed to players automatically once a week (every
Monday) from August 6, 2018, until June 10, 2019,
(44 weeks) by SMS and/or email with a direct link to an
online injury surveillance platform (AthleteMonitoring.-
com; FITSTATS Technologies). One school (n ¼35) started
registration on August 6, 1 school (n ¼33) on August 13, 2
schools (n ¼80) on September 3, and 2 schools (n ¼58) on
September 10. If players failed to complete the question-
naire, the system sent an automated reminder every day
until a response was received. Additionally, the principal
investigator sent SMS reminders to nonresponders after
3 and 5 days. The school physician and staff members
could access their players’ health information on a web-
based dashboard and encouraged players to respond. To
encourage participation, the principal investigator visited
3 of the 6 participating schools during December 2018
and January 2019 and maintained regular contact with
all players and responsible staff members throughout the
registration period.
OSTRC-H2 Questionnaire
The OSTRC-H2 consists of 4 questions about the athlete’s
participation in sports, modification in training or compe-
tition, performance, and symptoms of health problems dur-
ing the past 7 days.
16-18
If the athlete answered no on the
first questions (full participation without problems), the
questionnaire was complete for that week. If athletes
reported a health problem, they were asked about
*Address correspondence to Anine Nordstrøm, MD, Oslo Sports Trauma Research Center, Department of Sports Medicine, Norwegian School of Sports
Sciences, PB 4014 Ulleva
˚l Stadion 0806 Oslo, Norway (email: anine.nordstrom@nih.no).
†
Oslo Sports Trauma Research Center, Department of Sports Medicine, Norwegian School of Sports Sciences, Oslo, Norway.
‡
Sykehuset Innlandet HF, Innlandet Hospital Trust, Elverum, Norway.
§
Center for Disease Burden Norwegian Institute of Public Health, Bergen, Norway.
k
University of Oslo, Oslo University Hospital, Oslo, Norway.
Final revision submitted June 13, 2022; accepted July 24, 2022.
One or more of the authors has declared the following potential conflict of interest or source of funding: Financial support was received from the Innlandet
Hospital Trust (salary support for A.N.) and the Stiftelsen fondet til fremme av idrettsmedisin og idrettsfysioterapi. AOSSM checks author disclosures against
the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating
thereto.
Ethical approval for this study was obtained from the Norwegian Centre for Research Data (reference No. 59423 LH/LR) and the Norwegian Data
Inspectorate (reference No. 17/00803-6/CGN).
2Nordstrøm et al The Orthopaedic Journal of Sports Medicine
modification in training or competition, performance, and
symptoms and to define whether the health problem was an
injury or an illness. In case of an injury, they were asked to
classify it as either an acute injury (associated with a spe-
cific, clearly identifiable traumatic event) or an overuse
injury (no specific identifiable event responsible for the
occurrence) and register the affected anatomic area. The
players could not specify if their injuries were a result of
hockey participation. For all types of health problems, ath-
letes were asked to register the number of days of complete
time loss from training and competition (total inability to
train or compete) and whether the health problem had been
reported previously. They were asked to register all health
problems, and in cases of multiple problems the same week,
the questionnaire repeated itself.
Definition and Classification of Health Problems
We used an “all complaints” definition, recording all health
problems, irrespective of the need for medical attention or
the consequences for sports participation.
32,49,59
Our defini-
tions of health problem, acute injury, overuse injury, and
illness were consistent with the International Olympic
Committee consensus statement.
7
The definition of acute
injury was an injury caused by a single, clearly identifiable
energy transfer (eg, a fall or collision). The definition of
overuse injury was an injury caused by multiple accumula-
tive bouts of energy transfer without a single, clearly iden-
tifiable event responsible for the injury. Health problems
were defined as “substantial problems” if they caused mod-
erate or severe modifications to training, moderate or
severe reductions in performance, or a complete inability
to participate in ice hockey.
16-18
Load Data Collection
Additional questions about the number of training sessions
and games played in the previous 7 days were added to the
OSTRC-H2 questionnaire on October 29, 2018 (Table 1).
The players recorded the number of training sessions in
total, training sessions on ice, strength trainings, cardio-
vascular training, sessions with other type of activity,
games, games at the elite level, and shifts on ice at the elite
level.
Prevalence and Incidence Calculations
For each of the 33 weeks, we calculated the following
prevalence measures using the methods described by
Clarsen et al
17
: all health problems, substantial health
problems, all injuries, substantial injuries, all illnesses,
and substantial illnesses. We also calculated the average
weekly prevalence with 95%confidence intervals. The
incidenceofeachtypeofhealthproblemwasexpressed
as the number of cases per player per year (52 weeks).
The average time loss was expressed as days per athlete
per year (52 weeks).
Statistical Methods
All data were compiled using Microsoft Excel software
(Microsoft Office 365 ProPlus Version 2002). Data were
analyzed using Stata (StataCorp). The mean number of
training sessions and games per week were calculated
along with their standard deviations. Regarding the
training and game load data, 9%and 6%,respectively,
were missing; we assumed these to be missing at ran-
dom.
14
The data were imputed weekly with multiple
imputations using linear regression with chained equa-
tions.
3
The variables used in the imputation model were
age, player position, week, number of training sessions
and games in the 2 weeks before imputation, and the lat-
est number of training sessions and games (the variables
imputed). After each round of imputation, we calculated
the 2-week difference in training/game loads as well as
the cumulative training/game loads of the previous 2, 3,
4, and 6 weeks.
We used multivariable logistic regression with random
effects (mixed model) to investigate the association between
load variables and outcome measures.
51
For acute injuries,
we used data from the previous 2-week difference in train-
ing/game loads as well as the cumulative loads of the pre-
vious 2, 3, and 4 weeks. For overuse injuries, we used data
from the previous 2-week difference in training/game loads
as well as the cumulative loads of the previous 2, 4, and 6
weeks. The training and game load variables were modeled
separately to avoid multicollinearity. A random intercept
was included to account for within-player correlations. Age,
player position, week, and training sessions and games in
the current week were potential confounding factors
adjusted for. Week was modeled with a second-degree poly-
nomial term to account for potential nonlinearity.
4
In addi-
tion, to account for the exposure in the current week,
63
training sessions and games in the current week were
adjusted for. Models were run on each of 100 imputed data
sets, and results were pooled with Ruben’s rules. Statistical
significance was defined as P<.05.
TABLE 1
Custom Questions Added to the OSTRC-H2 Questionnaire
a
Exposure
How many games have you played the last 7 days?
Total number of games?
Total number of games on senior level?
How many shifts do you have on senior level?
How many training sessions have you had the last 7 days?
On ice
Strength training
Endurance training
Other activity/training
How many hours have you slept for the last 7 days?
Average
Night with least hours’ sleep
Night with most hours’ sleep
a
OSTRC-H2, Oslo Sports Trauma Research Center Overuse
Questionnaire on Health Problems.
The Orthopaedic Journal of Sports Medicine Training Load and Injury Risk in Elite Ice Hockey 3
RESULTS
Response Rate to the Weekly Questionnaires
We distributed 5223 questionnaires and received 4870
responses (average weekly response rate, 93%;range,
86%-99%) during the 33-week study period.
Number and Incidence of Health Problems
The players reported 133 acute injuries, 75 overuse inju-
ries, and 162 illnesses in total. This translated to 1.4 new
acute injuries, 0.8 new overuse injuries, and 1.7 new ill-
nesses per athlete per year (Table 2). The average time loss
was 34 days (95%CI, 33-35 days) per athlete per year,
17 days (95%CI, 16-19 days) for acute injuries, 8 days
(95%CI, 7-10 days) for overuse injuries, and 9 days
(95%CI, 7-10 days) for illnesses.
Prevalence of Health Problems
The average weekly prevalence of health problems, with
95%confidence intervals and ranges, is shown in Table 3.
Training and Game Load
The players reported an average of 8.8 (SD ±3.9) training
sessions and 0.9 (SD ±1.1) games per week (Table 4). The
weekly average number of training sessions fell slightly
throughout the registration period, and there were 2 peri-
ods with fewer training sessions: over a school holiday (reg-
istration weeks 9 and 10) and at the end of the hockey
season (registration weeks 24-28). The weekly average
number of games played was quite constant throughout the
registration period but had 2 periods with fewer games:
over a school holiday (registration weeks 8-10) and at the
end of the hockey season (registration weeks 24-26). No
games were played during weeks 27 to 33 of the registra-
tion. The number of training sessions and games by player
position and age is shown in Table 4.
Association of Training Sessions and Games to Injuries
Table 5 shows the association between training/game loads
and acute or overuse injuries, adjusted for player position,
age, week, second-degree polynomial term of the week, and
number of training sessions or games in the current week.
The relationship between cumulative number of training/
game loads and acute or overuse injuries had low effect sizes
and was not significant (OR, *1.0; P>.05 in all models)
(Table 5). The covariates of player position, age, week,
second-degree polynomial term of week, and number of
training sessions/games in the current week demonstrated
consistent results throughout all the models, for both acute
injuries and overuse injuries (Supplemental Table S1). None
of the participant characteristics (player position, age) were
significantly related to injury risk. Time was significant, as
well as the nonlinear term for time (week). Training and
game sessions for the current week had low odds ratios and
significant Pvalues. The random effects were significant in
all models (P<.05) (Supplemental Table S1). Analyses of
substantial acute and overuse injuries showed essentially
the same results as analyses of all acute and overuse injuries
(Table 5 and Supplemental Table S2).
DISCUSSION
This study is the first to investigate the association between
number of training sessions/games and injuries in junior
elite ice hockey players. We found no association between
TABLE 2
Number of Cases, Incidence, and Total Time Loss
of Acute Injuries, Overuse Injuries, and Illnesses
n
Incidence (cases per
athlete per year)
a
Total Time Lost, d
Acute injury 133 1.4 (1.2-1.7) 1601
Overuse injury 75 0.8 (0.6-1.0) 796
Illness 162 1.7 (1.5-2.0) 808
Total 370 4.0 (3.6-4.4) 3205
a
Incidence is shown with 95%CI.
TABLE 3
Average Weekly Prevalence of
All Health Problems and Substantial Health Problems
During the 33-Week Study Period
Weekly Prevalence, %
Variable Mean 95%CI Range
All health problems 23 21-25 12-34
Injuries 18 17-20 9-31
Acute injuries 9 8-10 3-17
Overuse injuries 10 9-10 5-15
Illness 6 5-6 1-11
Substantial health problems 15 14-16 7-23
Injuries 11 10-12 6-19
Acute injuries 7 6-8 2-13
Overuse injuries 5 4-5 3-9
Illness 4 4-5 1-8
TABLE 4
Training Sessions and Games by Player Position and Age
a
Training Sessions Games
Overall 8.8 ±3.9
(range, 0-23)
0.9 ±1.1
(range, 0-7)
Player position
Defenders (n ¼63) 8.7 ±3.6 0.9 ±1.2
Forwards (n ¼76) 8.6 ±4.1 1.0 ±1.2
Goalkeepers (n ¼20) 9.9 ±4.2 0.6 ±0.9
Player age, y
15 (n ¼13) 10.0 ±3.9 0.9 ±1.1
16 (n ¼67) 9.2 ±4.0 0.9 ±1.1
17 (n ¼50) 8.2 ±4.0 0.9 ±1.1
18 (n ¼29) 8.4 ±3.9 1.1 ±1.2
a
Data are shown as mean ±SD unless otherwise indicated.
4Nordstrøm et al The Orthopaedic Journal of Sports Medicine
the cumulative long-term or 2-week difference in the num-
ber of training sessions or games and the risk of acute or
overuse injuries.
Developing youth athletes are at increased risk of injury
when introduced to new loads, changes in loads, or con-
gested competition calendars.
11,15,25,30
The most talented
athletes might be more prone to injury,
6
andloadvol-
ume
54,61
and impact load
54,62
are believed to represent
important risk factors for injury. To our knowledge, there
is no previous study in junior ice hockey reporting on train-
ing and game load. The players reported an average of 8.8
training sessions and 0.9 games per week, reflecting their
status as elite junior players. This training volume is sim-
ilar to that of elite youth handball.
12
Other studies have
reported on the training load in football,
15,22
but differences
in quantifying workload make direct comparisons to our
findings difficult.
The ACWR has recently been criticized because of
methodological issues, and the validity of the model has
been questioned.
38-40
Methodological choice affects the
potential association between ACWR and health pro-
blems
21
; however, the causal relation to injury has not been
established.
40
The proposed relationship between ACWR
and health problems is based on descriptive studies
reporting associations between various alterations of
ACWR and health problems. Several studies have exam-
ined the relationship between training load and health pro-
blems in junior sports.
5,15,22,23,42,48,56
Most of these have
used the ACWR, but none have used the same calculation
of ACWR, analytical approach, or statistical methods.
Despite these limitations, studies in football
15,23
and other
sports
42,48
have found that variations in the ACWR were
associated with the risk of health problems. In contrast, 1
study based on the ACWR found that internal load markers
were not associated with noncontact injuries in young foot-
ball players.
56
The only randomized trial performed (482
elite youth soccer players of both sexes) using the ACWR
to manage player load within what was believed to repre-
sent the “safe zone” found no between-group difference in
health problem prevalence, suggesting that the specific
load management intervention was not successful in pre-
venting health problems.
22
Because of the recent criticism of the ACWR concept,
we chose not to use ACWR for our analyses. A priori, we
decided to use measures likely to be most valid and clini-
cally relevant. It is not known how many weeks back in
time training and game load can have an impact on injury,
and this may differ between sports. Our data only included
TABLE 5
Model Results From Number of Training Sessions/Games and All Injuries and Substantial Injuries
a
Cumulative Load
2-Week Difference 2 Weeks 3 Weeks 4 Weeks 6 Weeks
OR (95%CI) POR (95%CI) POR (95%CI) POR (95%CI) POR (95%CI) P
All Injuries
Training sessions
b
Acute injury (n ¼451) 0.99
(0.96-1.03)
.68 1.01
(0.98-1.03)
.61 1.00
(0.97-1.02)
.69 1.00
(0.98-1.02)
.76
Overuse injury (n ¼468) 1.00
(0.96-1.05)
.99 1.00
(0.98-1.02)
.79 0.99
(0.98-1.01)
0.58
Games
b
Acute injury (n ¼451) 1.03
(0.91-1.17)
.60 0.97
(0.89-1.06)
.49 0.97
(0.90-1.04)
.40 0.98
(0.92-1.04)
.48
Overuse injury (n ¼468) 0.98
(0.84-1.15)
.79 0.97
(0.88-1.05)
.43 0.98
(0.91-1.06)
.57
Substantial Injuries
Training sessions
b
Acute injury (n ¼339) 1.00
(0.95-1.04)
.90 1.00
(0.97-1.03)
.86 0.99
(0.97-1.02)
.69 1.00
(0.98-1.02)
.84
Overuse injury (n ¼220) 1.02
(0.97-1.07)
.53 1.01
(0.98-1.03)
.54 0.99
(0.97-1.02)
.55
Games
b
Acute injury (n ¼339) 0.98
(0.84-1.13)
.74 0.98
(0.88-1.09)
.77 0.96
(0.88-1.05)
.38 0.97
(0.90-1.05)
.49
Overuse injury (n ¼220) 0.97
(0.80-1.19)
.79 0.93
(0.84-1.04)
.21 0.93
(0.85-1.02)
.12
a
Data are based on logistic mixed models adjusted for player position, age, week, and number of training sessions and games in the week of
data registration (current exposure).
b
Sample sizes denote the total number of weeks with a reported injury.
The Orthopaedic Journal of Sports Medicine Training Load and Injury Risk in Elite Ice Hockey 5
the total number of training sessions and games; we do not
have information about their intensity or duration. It was
not possible to investigate current exposure without bias as
we only recorded load once a week and did not register
exactly when during the week an injury occurred. However,
this is not the case for previous cumulative exposure. Direct
comparison of studies using ACWR as load measures to our
findings is therefore difficult.
In a small study in collegiate ice hockey (1 team, 26
players), Mustapich and Koehle
50
found that players with
greater 2-day training load experienced greater odds of
injury compared with athletes with lower loads. This study
defined injury as that related to ice hockey play, resulting
in time loss or modification of a training session or game,
diagnosed by the team physician or physical therapist. The
subjective training load for all on-ice sessions was recorded
using the subjective rating of perceived exertion (modified
Borg CR-10 RPE). Each athlete’s score was multiplied with
the duration in training or game sessions in minutes, deter-
mining the session training load value in arbitrary units.
They used cumulative training load (each day, 2 days, and 2
weeks) measured in arbitrary units. Cumulative training
load over 2 weeks affected the odds of injury occurrence
positively, but this finding was not significant. These find-
ings are not in line with ours, although direct comparisons
are difficult, as we separated training from game load, and
acute from overuse injuries, and only recorded the number
of training sessions and games.
We found no association between the number of training
sessions/games and injuries. An explanation for this may be
that the relationship between training load and health pro-
blems is complex, and several other factors might influence
the occurrence of health problems: previous injury status,
fitness, wellness, nonsporting load, biomechanics, poor
technique, differences in reporting of injuries, differences
in pain threshold and pain tolerance, or simply chance.
8
Meeuwisse
46
and Meeuwisse et al
47
have previously dem-
onstrated how internal and external risk factors can influ-
ence injury risk. Later, Bahr and Holme
8
and Bahr and
Krosshaug
9
expanded this conceptual model and consid-
ered how injury mechanisms—the inciting event—also may
play a crucial role. Further, cumulative long-term exposure
might have a small-to-moderate association to injury, and
our sample size might not have been large enough to detect
small-to-moderate associations.
8
Despite conflicting evidence, training load is considered
to be an important risk factor for health problems. Previous
research has suggested that the relationship between
training load and injury risk should be assumed to be non-
linear.
4
However, future research on the methods that are
most optimal to determine the relationship between train-
ing load and risk of injury is needed.
Methodological Considerations
One of the main strengths of the study is the high response
rate and a relatively large sample size. We also used sensitive
injury surveillance methods to capture all health problems
and a multivariable statistical approach to investigate train-
ing and game load as potential risk factors for injury.
This study also has several limitations. The data were
collected from the players, and the extent to which health
problems and numbers of trainings sessions and games
were underreported could not be measured. The weekly
reports by the athletes are subjective, and the reporting
threshold may differ between players. Recall bias and
underreporting of health problems and training sessions
and games could also affect the results; daily reports could
reduce this bias but, on the other hand, challenge the com-
pliance of the participants.
The players reported the number of training sessions in
total, training sessions on ice, strength trainings, cardio
trainings, sessions with other types of activity, games, and
games and shifts on ice at the elite level. In our analyses,
we only used the total number of training sessions and
games; we did not ask for information about their inten-
sity or duration. However, the duration of a training ses-
sion was very consistent, since access to time on ice was
limited; typically, 60 to 75 minutes was allotted to each
team for each session. Strength sessions typically lasted
45 to 60 minutes. We would argue that the number of
training sessions is an adequate marker for training load;
adding duration in minutes would have minimal impact on
the outcome.
Still, the intensity of training sessions and games may
vary substantially, even between players in the same ses-
sion. We therefore would have liked to have had data on
both external (eg, GPS data) and internal (eg, session rat-
ing of perceived exertion) loads. This has been done in other
sports, and we recommend this for future studies on train-
ing and game load in ice hockey.
While this study included 159 players monitored for
33 weeks, resulting in nearly 5000 observations, sample
size is still a limitation and may not be sufficient to detect
small-to-moderate associations between training/game
load and injury. Nevertheless, across all models, the odds
ratio is centered around 1, with narrow confidence inter-
vals. This consistent finding indicates that the risk of a type
2 error is low and that adding more observations is unlikely
to change our conclusions.
Health-related problems are expected in ice hockey. The
wide definition used, based on all health complaints, leads
to the registration of minor and transient problems (eg,
muscle soreness or unspecific symptoms).
17
This is a
source of systematic bias, overestimating the prevalence
of sports-related health problems. The “substantial health
problem” definition (problems leading to reduced perfor-
mance and/or participation) might be a more appropriate
estimate of the impact of injuries and illnesses in ice
hockey.
Finally, our study only included male junior elite ice hockey
players and may not be generalizable to other populations.
CONCLUSION
In the current study of junior elite ice hockey players, there
was no evidence of an association between cumulative expo-
sure to training or game loads and injury risk.
6Nordstrøm et al The Orthopaedic Journal of Sports Medicine
ACKNOWLEDGMENT
The authors thank all the athletes for their participation,
as well as the schools, clubs, and medical teams involved.
They thank the Innlandet Hospital Trust and Stiftelsen
fondet til fremme av idrettsmedisin og idrettsfysioterapi for
financial support. They also thank Lien My Diep and Har-
ald Weedon-Fekjær for excellent help with multiple impu-
tation. The Oslo Sports Trauma Research Center has been
established at the Norwegian School of Sport Sciences
through generous grants from the Royal Norwegian Minis-
try of Culture, the South-Eastern Health Authority, the
International Olympic Committee, the Norwegian Olympic
and Paralympic Committee and Confederation of Sports,
Norsk Tipping, and Sparebankstiftelsen DnB.
REFERENCES
1. Akenhead R, Nassis GP. Training load and player monitoring in high-
level football: current practice and perceptions. Int J Sports Physiol
Perform. 2016;11(5):587-593.
2. Anderson L, Triplett-McBride T, Foster C, Doberstein S, Brice G.
Impact of training patterns on incidence of illness and injury during
a women’s collegiate basketball season. J Strength Cond Res. 2003;
17(4):734-738.
3. Bache-Mathiesen LK, Andersen TE, Clarsen B, Fagerland MW.
Handling and reporting missing data in training load and injury
risk research. Sci Med Footb. Published online November 17,
2021. doi:10.1080/24733938.2021.1998587
4. Bache-Mathiesen LK, Andersen TE, Dalen-Lorentsen T, Clarsen B,
Fagerland MW. Not straightforward: modelling non-linearity in training
load and injury research. BMJ Open Sport Exerc Med. 2021;7(3):
e001119.
5. Bacon CS, Mauger AR. Prediction of overuse injuries in professional
U18-U21 footballers using metrics of training distance and intensity.
J Strength Cond Res. 2017;31(11):3067-3076.
6. Bahr R. Demise of the fittest: are we destroying our biggest talents? Br
J Sports Med. 2014;48(17):1265-1267.
7. Bahr R, Clarsen B, Derman W, et al. International Olympic Committee
consensus statement: methods for recording and reporting of epide-
miological data on injury and illness in sport 2020 (including STROBE
Extension for Sport Injury and Illness Surveillance (STROBE-SIIS)).
Br J Sports Med. 2020;54(7):372-389.
8. Bahr R, Holme I. Risk factors for sports injuries—a methodological
approach. Br J Sports Med. 2003;37(5):384-392.
9. Bahr R, Krosshaug T. Understanding injury mechanisms: a key com-
ponent of preventing injuries in sport. Br J Sports Med. 2005;39(6):
324-329.
10. Bengtsson H, Ekstrand J, Hagglund M. Muscle injury rates in profes-
sional football increase with fixture congestion: an 11-year follow-up
of the UEFA Champions League injury study. Br J Sports Med. 2013;
47(12):743-747.
11. Bergeron MF, Mountjoy M, Armstrong N, et al. International Olympic
Committee consensus statement on youth athletic development. Br J
Sports Med. 2015;49(13):843-851.
12. Bjørndal CT, Bache-Mathiesen LK, Gjesdal S, Moseid CH, Myklebust
G, Luteberget LS. An examination of training load, match activities,
and health problems in Norwegian youth elite handball players over
one competitive season. Front Sports Act Living. 2021;3:635103.
13. Blanch P, Gabbett TJ. Has the athlete trained enough to return to play
safely? The acute:chronic workload ratio permits clinicians to quantify
a player’s risk of subsequent injury. Br J Sports Med. 2016;50(8):
471-475.
14. Borg DN, Nguyen R, Tierney NJ. Missing data: current practice in
football research and recommendations for improvement. Sci Med
Footb. 2022;6(2):262-267.
15. Bowen L, Gross AS, Gimpel M, Li FX. Accumulated workloads and the
acute:chronic workload ratio relate to injury risk in elite youth football
players. Br J Sports Med. 2017;51(5):452-459.
16. Clarsen B, Bahr R, Myklebust G, et al. Improved reporting of overuse
injuries and health problems in sport: an update of the Oslo Sport
Trauma Research Center questionnaires. Br J Sports Med. 2020;
54(7):390-396.
17. Clarsen B, Myklebust G, Bahr R. Development and validation of a new
method for the registration of overuse injuries in sports injury epide-
miology: the Oslo Sports Trauma Research Centre (OSTRC) overuse
injury questionnaire. Br J Sports Med. 2013;47(8):495-502.
18. Clarsen B, Ronsen O, Myklebust G, Florenes TW, Bahr R. The Oslo
Sports Trauma Research Center questionnaire on health problems: a
new approach to prospective monitoring of illness and injury in elite
athletes. Br J Sports Med. 2014;48(9):754-760.
19. Colby MJ, Dawson B, Heasman J, Rogalski B, Gabbett TJ. Acceler-
ometer and GPS-derived running loads and injury risk in elite Austra-
lian footballers. J Strength Cond Res. 2014;28(8):2244-2252.
20. Cross MJ, Williams S, Trewartha G, Kemp SP, Stokes KA. The influ-
ence of in-season training loads on injury risk in professional rugby
union. Int J Sports Physiol Perform. 2016;11(3):350-355.
21. Dalen-Lorentsen T, Andersen TE, Bjørneboe J, et al. A cherry, ripe for
picking: the relationship between the acute-chronic workload ratio
and health problems. J Orthop Sports Phys Ther. 2021;51(4):162-173.
22. Dalen-Lorentsen T, Bjørneboe J, Clarsen B, Vagle M, Fagerland MW,
Andersen TE. Does load management using the acute:chronic work-
load ratio prevent health problems? A cluster randomised trial of 482
elite youth footballers of both sexes. Br J Sports Med. 2021;55(2):
108-114.
23. Delecroix B, Delaval B, Dawson B, Berthoin S, Dupont G. Workload
and injury incidence in elite football academy players. J Sports Sci.
2019;37(24):2768-2773.
24. Delecroix B, McCall A, Dawson B, Berthoin S, Dupont G. Workload
and non-contact injury incidence in elite football players competing in
European leagues. Eur J Sport Sci. 2018;18(9):1280-1287.
25. DiFiori JP, Benjamin HJ, Brenner JS, et al. Overuse injuries and burn-
out in youth sports: a position statement from the American Medical
Society for Sports Medicine. Br J Sports Med. 2014;48(4):287-288.
26. Drew MK, Finch CF. The relationship between training load and injury,
illness and soreness: a systematic and literature review. Sports Med.
2016;46(6):861-883.
27. Dupont G, Nedelec M, McCall A, McCormack D, Berthoin S, Wisloff
U. Effect of 2 soccer matches in a week on physical performance and
injury rate. Am J Sports Med. 2010;38(9):1752-1758.
28. Eckard TG, Padua DA, Hearn DW, Pexa BS, Frank BS. The relation-
ship between training load and injury in athletes: a systematic review.
Sports Med. 2018;48(8):1929-1961.
29. Fanchini M, Rampinini E, Riggio M, Coutts AJ, Pecci C, McCall A.
Despite association, the acute:chronic work load ratio does not pre-
dict non-contact injury in elite footballers. Sci Med Footb. 2018;2(2):
108-114.
30. Fortington LV, Berry J, Buttifant D, Ullah S, Diamantopoulou K, Finch
CF. Shorter time to first injury in first year professional football players:
a cross-club comparison in the Australian Football League. J Sci Med
Sport. 2016;19(1):18-23.
31. Foster C. Monitoring training in athletes with reference to overtraining
syndrome. Med Sci Sports Exerc. 1998;30(7):1164-1168.
32. Fuller CW, Ekstrand J, Junge A, et al. Consensus statement on injury
definitions and data collection procedures in studies of football (soc-
cer) injuries. Clin J Sport Med. 2006;16(2):97-106.
Supplemental material for this article is available at
https://journals.sagepub.com/doi/full/10.1177/23
259671221129646#supplementary-materials.
The Orthopaedic Journal of Sports Medicine Training Load and Injury Risk in Elite Ice Hockey 7
33. Gabbett TJ. Influence of training and match intensity on injuries in
rugby league. J Sport Sci. 2004;22(5):409-417.
34. Gabbett TJ. The training—injury prevention paradox: should athletes
be training smarter and harder? Br J Sports Med. 2016;50(5):273-280.
35. Hulin BT, Gabbett TJ, Blanch P, Chapman P, Bailey D, Orchard JW.
Spikes in acute workload are associated with increased injury risk in
elite cricket fast bowlers. Br J Sports Med. 2014;48(8):708-712.
36. Hulin BT, Gabbett TJ, Lawson DW, Caputi P, Sampson JA. The acute:
chronic workload ratio predicts injury: high chronic workload may
decrease injury risk in elite rugby league players. Br J Sports Med.
2016;50(4):231-236.
37. Ha
¨gglund M, Wald´
en M, Magnusson H, Kristenson K, Bengtsson H,
Ekstrand J. Injuries affect team performance negatively in profes-
sional football: an 11-year follow-up of the UEFA Champions League
injury study. Br J Sports Med. 2013;47(12):738-742.
38. Impellizzeri F, Woodcock S, Coutts A, Fanchini M, McCall A, Vigotsky
A. Acute to random workload ratio is ‘as’ associated with injury as
acute to actual chronic workload ratio: time to dismiss ACWR and its
components. SportRxiv. Preprint posted March 3, 2020. doi:10.31236/
osf.io/e8kt4
39. Impellizzeri F, Woodcock S, McCall A, Ward P, Coutts A. The acute-
chronic workload ratio-injury figure and its ‘sweet spot’ are flawed.
SportRxiv. Preprint posted June 3, 2019. doi:10.31236/osf.io/gs8yu
40. Impellizzeri FM, Tenan MS, Kempton T, Novak A, Coutts AJ. Acute:
chronic workload ratio: conceptual issues and fundamental pitfalls.
Int J Sports Physiol. Published online June 5, 2020. doi:10.1123/ijspp.
2019-0864
41. Jaspers A, Kuyvenhoven JP, Staes F, Frencken WGP, Helsen WF,
Brink MS. Examination of the external and internal load indicators’
association with overuse injuries in professional soccer players. J Sci
Med Sport. 2018;21(6):579-585.
42. Lathlean TJH, Gastin PB, Newstead SV, Finch CF. Absolute and rel-
ative load and injury in elite junior Australian football players over
1 season. Int J Sports Physiol. 2019;15(4):511-519.
43. Lolli L, Batterham AM, Hawkins R, et al. The acute-to-chronic work-
load ratio: an inaccurate scaling index for an unnecessary normal-
isation process? Br J Sports Med. 2019;53(24):1510-1512.
44. Malone S, Owen A, Newton M, Mendes B, Collins KD, Gabbett TJ.
The acute:chronic workload ratio in relation to injury risk in profes-
sional soccer. J Sci Med Sport. 2017;20(6):561-565.
45. McKnight CM, Ferrara MS, Czerwinska JM. Intercollegiate ice hockey
injuries: a three-year analysis. J Atht Train. 1992;27(4):338-343.
46. Meeuwisse W. Assessing causation in sport injury: a multifactoral
model. Clin J Sport Med. 1994;1994(4):166-170.
47. Meeuwisse WH, Tyreman H, Hagel B, Emery C. A dynamic model of
etiology in sport injury: the recursive nature of risk and causation. Clin
J Sport Med. 2007;17:215-219.
48. Moreno-P ´
erez V, Prieto J, Del Coso J, et al. Association of acute and
chronic workloads with injury risk in high-performance junior tennis
players. Eur J Sport Sci. 2021;21(8):1215-1223.
49. Mountjoy M, Junge A, Alonso JM, et al. Consensus statement on the
methodology of injury and illness surveillance in FINA (aquatic sports).
Br J Sports Med. 2016;50(10):590-596.
50. Mustapich C, Koehle M. The effects of training load and non-training
stress on injury risk in collegiate ice hockey players. Transl Sports
Med. 2021;4(6):931-936.
51. Nielsen RØ, Shrier I, Casals M, et al. Statement on methods in sport
injury research from the first METHODS MATTER meeting, Copenha-
gen, 2019. Br J Sports Med. 2020;50(5):226-233.
52. Nordstrøm A, Bahr R, Clarsen B, Talsnes O. Prevalence and burden of
self-reported health problems in junior male elite ice hockey players: a
44-week prospective cohort study. Am J Sports Med. 2021;49(12):
3379-3385.
53. Nordstrøm A, Bahr R, Talsnes O, Clarsen B. Prevalence and burden of
health problems in male elite ice hockey players: a prospective study
in the Norwegian professional league. Orthop J Sports Med. 2020;
8(2):2325967120902407.
54. Olsen SJ II, Fleisig GS, Dun S, Loftice J, Andrews JR. Risk factors for
shoulder and elbow injuries in adolescent baseball pitchers. Am J
Sports Med. 2006;34(6):905-912.
55. Piggott B, Newton M, McGuigan M. The relationship between
training load and incidence of injury and illness over a pre-
season at an Australian Football League club. JAustStrength
Cond. 2009;17:4-17.
56. Raya-Gonz ´
alez J, Nakamura FY, Castillo D, Yanci J, Fanchini M.
Determining the relationship between internal load markers and non-
contact injuries in young elite soccer players. Int J Sports Physiol.
2019;14(4):421-425.
57. Smith AM, Stuart MJ, Wiese-Bjornstal DM, Gunnon C.Predictors of
injury in ice hockey players. A multivariate, multidisciplinary
approach. Am J Sports Med. 1997;25(4):500-507.
58. Stuart MJ, Smith A.Injuries in Junior A ice hockey. A three-year
prospective study. Am J Sports Med. 1995;23(4):458-461.
59. Timpka T, Alonso JM, Jacobsson J, et al. Injury and illness definitions
and data collection procedures for use in epidemiological studies in
athletics (track and field): consensus statement. Br J Sports Med.
2014;48(7):483-490.
60. Tuominen M, Stuart MJ, Aubry M, Kannus P, Parkkari J. Injuries in
world junior ice hockey championships between 2006 and 2015. Br J
Sports Med. 2017;51(1):36-43.
61. Visnes H, Aandahl HA, Bahr R. Jumper’s knee paradox—jumping
ability is a risk factor for developing jumper’s knee: a 5-year prospec-
tive study. Br J Sports Med. 2013;47(8):503-507.
62. Visnes H, Bahr R. Training volume and body composition as risk
factors for developing jumper’s knee among young elite volleyball
players. Scand J Med Sci Sports. 2013;23(5):607-613.
63. Wang C, Vargas JT, Stokes T, Steele R, Shrier I. Analyzing activity and
injury: lessons learned from the acute:chronic workload ratio. Sports
Med. 2020;50(7):1243-1254.
8Nordstrøm et al The Orthopaedic Journal of Sports Medicine
